Robust and fragile watermarking for medical imaging
Bowen Meng, Xiaochen Yuan · 2025
In the rapidly evolving landscape of smart healthcare systems, the secure transmission and robust integrity of Electronic Patient Records (EPR) are of utmost importance. Digital watermarking has been a reliable method for preserving the authenticity of digital images, but the challenge of ensuring the security and confidentiality of patient data in medical imaging persists. This paper introduces a data hiding scheme for medical images, incorporating a fragile watermarking technique using Least Significant Bit (LSB), as well as robust watermarking using Discrete Wavelet Transform (DWT). The intensity-based image segmentation method is employed to select the Region of Non-Interest (RONI) for embedding the data and watermark. The unique contribution of this research is the application of both fragile and robust watermarking techniques in one image while maintaining losslessness in the Region of Interest (ROI), enabling tamper detection. Experiments conducted on The Cancer Genome Atlas Lung Adenocarcinoma Collection (TCGA-LUAD) demonstrate the improved performance of the proposed method in terms of security, authenticity, and integrity while maintaining the quality of medical images. This research represents a significant stride towards a more secure and efficient approach to safeguarding patient data in medical imaging.